100,000 people dead each year due to medical mistakes? Really?
 

Apparently Dennis Quaid is going to be on 60 Minutes talking about his twin infants' horrible ordeal at a hospital that administered the wrong medicine to them.

In the process he says 100,000 Americans die in hospitals each year from medical mistakes.

Is it me, or does this sound a little high?

http://www.cbsnews.com/stories/2008/...n3936412.shtml

The number I have heard is 10,000. Maybe he got confused. But maybe one of our board docs can set us straight.

100k is the number I have heard.

We are always trying to implement new procedures. Problems include bad handwriting, medications that sound or are spelled similarly, and many other things.

For example, I had a friend who told me once he ordered 4mg Magnesium Sulfate to be given IV. The nurse called him back said "We don't have 4mg of morphine, I've checked". And he almost shat a brick.

In case you don't know, 4mg of morphine IV would probably kill a horse.

Even if it looked like he wrote for morphine, the nurse should have known not administer that much.

My wife was given a mag wash while on bedrest in the hospital. It caused some sort of hallucinogenic side effects. But it cut down on the contractions. I think she slept for almost an entire day on that stuff.

It depends on the definition of medical mistake.

Unfortunately, most people expect docs to be perfect, which is silly.

I've seen examples of people who are dying of cancer to undergo a procedure where a mistake is made that causes death to occur earlier than it otherwise would. So there is a medical mistake that caused death in a person who was going to die soon anyway.

I also know of an example in Utah where a student had appendicitis and the anesthesiologist accidentally injected him with insulin instead of an anesthetic because the bottles looked identical. (The previous patient had been a diabetic so they had insulin on the table to help control that patient's blood sugar.) The insulin killed the poor kid.

Or you can have a psych patient who cheeks his pills in the psych ward and then ODs on them days later, and then the patient's family sues the doctor for giving him the pills even though they were needed.

So it's a question of where you draw the line. Some things are obvious medical mistakes (like the insulin thing), whereas the psych patient who cheeks his pills and ODs is probably not a mistake, but some might classify it as such.

For some reason, I thought you were a doctor.

No...4 mg of morphine would not kill a horse. It's probably the most common dose of pain medicine I write as the morphine vials come in either 4 mg or 10 mg.

And I would like to see someone measure out 4 mg of magnesium. There's probably more than that on your fingertip right now.

I'm thinking your friend 4 g of morphine--which would kill a horse--as 4 g is a normal dose of magnesium.

BUT, there's no one in the world that would give 4 g of morphine. I doubt you
could even find that much in the pharmacy.

This morphine vs magnesium is one of the justifications for the ridiculous JCAHO requirements of writing out morphine and magnesium, but the truth is, I seriously doubt this mistake has ever been made in the history of medicine. Another set of requirements by someone who's never treated a patient.

Oops, you are right. Meant 4 grams.

I've provided you with an example of how a patient could have been killed and you call it ridiculous, in the face of 100,000 deaths a year.

You are a bigger doofus than I thought.

Yes, the 100,000 number is the number in the IOM study done recently. However, there are serious methodologic issues to that study, not the least of which is causality. Any time there was a death in a case where a mistake was made, the death was attributed to that mistake, regardless of causality. I think harm-causing mistakes are actually much rarer than that, and I'm not alone in thinking that.

It goes back to an old standby of mine--if something sounds wrong, it probably is. Very few surprising statistics can not be explained away rather easily.

Thanks, Dennis.

Good analysis (and examples).

Although, I have to say, the insulin thing is sort of surprising for a few reasons. I've had patients take megadoses of insulin (up to 100x their dose); death from this would be unusual.

Which goes to my original point on this study--it's actually fairly hard to kill a patient in the present system. Sure, I see mistakes made all of the time--but most of them are either: 1) very minor (with no real consequences), 2) caught at some point in the process (pharmacy errors), 3) consequences fixed before permanent damage is done, or 4) done in the heat of the moment (where the patient's dying anyway, and it's VERY questionable whether the mistake caused the death). No way that 100,000 people are killed each year by medical mistakes.